1. Field of the Invention
The present invention relates to an image processing apparatus for compressing and processing image data for a facsimile system or an electronic file system.
2. Description of the Related Art
In a conventional facsimile system or electronic file system, image information is compressed in order to improve transmission efficiency and storage efficiency, and the compressed image information is transmitted and stored.
Typical compression methods are a modified Huffman (MH) method and a modified READ (MR) method. In these methods even if an error occurs in compressed data, only the data portion causing an error is lost, and the error does not influence other data (the entire image), but this is at the cost of compression efficiency.
Recently, with the use of high-quality digital transmission paths or error-free optical disc units, the probability of error generation in compressed data tends to be decreased. With this tendency, another compression method, e.g., a modified modified READ (MMR) method capable of achieving a higher compression ration than the MH or MR method is often employed.
In the high-compression MMR method, compression at a high ratio 1.5 to 3.0 times that obtained with the MH method can be achieved. However, if an error occurs in the compressed data, the entire image following the data portion containing the error is influenced by the error, and hence, it is impossible to reproduce the image properly.
In another proposal, error-check-and-correct (ECC) data is added to data compressed by the MMR method to realize error-free data.
However, if a compression circuit malfunctions due to high-compression processing, erroneous compression operation is performed, and erroneously compressed data is transmitted and stored, then the compression operation error can be detected only after the compressed data is expanded.
Therefore, when a long time interval is present between image data compression processing and expansion processing of compressed image data like in a facsimile system having a storage function or an image electronic file system, if a compression error is detected upon expansion processing, original data has already been deleted, and information may be lost.